SOME

REFLECTIONS
ON THE
MAGIC-RELIGION
H.S.

RELATIONSHIP

VERSNEL

Summary
of magic as distinct
The well-known substantialist-'Frazerian'-definitions
from religion by its immediate and individual goals, the concomitant manipulative
and coercive attitude, the instrumental and mechanical type of action etc., have
been under attack for more than half a century. Anthropologists in particular have
argued that no meaningful contrast between religion and magic can be gained
from this approach and that our notion 'magic' is a modern-western biased construct which does not fit representations of other cultures. Consequently, in the
view of some of them, the term 'magic' should be altogether avoided. Furthermore, with respect to the ancient and early modern world, in which the opposition
religion-magic is supposed to have originated, it is argued that magic and religion
function exclusively
as value-judgments, terms indicating 'magic' being exploited to
stigmatize illegitimate or undesired (religious) behaviour of socially or culturally
deviant groups.
In the present article it is argued that-although admittedly this functionalist
approach has yielded remarkable and lasting results-rejection of the term 'magic'
will soon turn out to be unworkable and, in fact, is putting the cart before the
horse. From an etic point of view-which in the view of the author is the only
possible way to conduct scholarly discourse-it will be impossible to do cultural
research without the aid of heuristic instruments such as-at least broad,
polythetic or prototypical-definitions. And, if possible at all, it would be utterly
unpractical to completely eliminate religion as one of the obvious models of
contrast.
This position is substantiated with some practical instances from the GraecoRoman world. It is shown that, at least in the context of (magical) curse-tablets
and-related but clearly distinct-(religious) prayers for justice or vengeance, the
ancient authors were clearly aware of the very same distinctions modern people
normally associate with the notions of magic and religion.
Definitions and their critics
Magic does not exist, nor does religion. What do exist are our
of these concepts.'
Scholars in earlier decades of this
definitions
century were luckier: they knew both what magic was and how to
to those of either
find it. They simply opposed its characteristics
science or religion, which they knew as well. It all began with
Frazer's tripartite distinction between science, religion and magic,

178
the first being defined as knowledge which is actually or potentially
the second and third belonging
to the
verifiable
by research,
which
is
asserted
of
the
truth
of
dogmatic
knowledge,
category
In his view, the basic difwithout regard to empirical verification.
and
ference between the latter two was that religion is intrinsically
if
is
but
to
an
whereas
not,
will,
magic
put
essentially unverifiable,
or
as "bastard
science"
test, prove false. Magic,
empirical
art", and science proper alike claim to offer opera"pretended
tional knowledge which can be exploited to attain control over the
environment
and to achieve concrete goals directly, without mediaand
tion. Magic
religion have in common that they refer to supernatural forces and powers, a reality different from normal reality.
It will be evident that there is a tension between the notion of nonmediation
(that is: without needing a personal medium other than
course not without instrumental
the magician or his client-of
aid)
and that of the supernatural
agent, a tension which was bound to
discussions.
surface in subsequent
In the framework
of these Frazerian
definitions
of religion and
will
be
on
which
is
silent
the
outside
science,
magic-I
scope of my
issue2-a
set of distinctions
has been elaborated
which have
dominated
the discussion
in the
years after their formalization
of 11 oppositions
much-quoted
catalogue
by the anthropologist
Goode.3
the serious criticism
that had already
been
Though
in
the
has
been
thirties
and
elaborated
with
ever
expressed
repeated
in
recent
the
distinctionsdecades,
increasing emphasis
following
a selection of the most important
items-are
still generally applied,
at any rate outside the field of anthropology.
to achieve concrete,
1 ) Intention. Magic is employed
mostly
individual goals. Religion is not primarily purpose-motivated,
or at
most focusses on intangible long-term
which
concern
collecgoals
tive issues of society.4 4
Man is both the
2) Attitude. Magic is essentially manipulative.
initiator and the executor of processes he controls with the aid of
knowledge which he has, or which is put, at his disposal. Religion
views man as dependent
upon powers outside his sphere of influence. This entails an attitude of submission
and supplication.
The
coercive manipulaopposition is thus one between "instrumental,
tion" and "personal,
supplicative
negotiation."

179
by the attention
3) Action. Magic is characterized
paid to the
technical side of the manipulation,
of
formula
and modus
precision
is often required
since the
operandi. Professional
experience
are observed, there
knowledge is secret. But if all the instructions
is an expectation
of direct results. In so far as religion, on the other
admits
of
effects (prayer for health, votives, private
intended
hand,
the results are never dependent
upon a professional
oracles),
factor,
specialist, though his skill may be required as a mediating
nor on the suppliant,
but solely and exclusively on the free favour
of sovereign gods.
4) Social/moral evaluation. Since the goals of magic often run
of the society, magic
counter to the interests of other members
the
of
an
anti-social
or at least a-social
connotation
acquires
easily
to
the
Durkheimian
thus
leading
dichotomy:
magic is
activity,
whereas
has
immoral,
deviant,
anti-social,
positive social
religion
is
cohesive
and
functions,
solidarizing.
One might, of course, elaborate upon distinctions
but these suffice for our purpose. In order to illustrate their influence I note that
all, except the last, are implied in the definition given by no less an
than Sir Edmund Leach in 1964: "The term magic
anthropologist
denotes a complex of belief and action on the basis and by means
of which persons and groups may attempt to control their environment in such a way as to achieve their ends, the efficacy of such
control being untested and in some cases untestable by the methods
of empirical science. The core of the magical act is that it rests on
untested belief and that it is an effort at control. The
empirically
first aspect
the second
from
it from
science,
distinguishes
5
religion. "5
no item in this scheme has escaped criticism. It has
Practically
been demonstrated
that typically
"mechanical"
(ex opere operato)
devices lie at the heart of religious
magical
(ex opere operantis)
most revealingly
in the grand debate between Proceremonies,
testants and Roman Catholics in the 16th century.6 As far as antithe typical rites of primitive
Roman religion
quity is concerned,
have induced
its characterization
as a "magical
Vice
religion".
versa, some expressions of the magical papyri of late antiquity sometimes cannot be distinguished
from religious confession. It has also

180
been shown that magic can have social goals (for instance collective
and
spells and devices for a good harvest or against catastrophes)
destructive
forms
of
religion a-social effects (hermits, sectarianism,
Again, that people do not always display more belief
crisis-cults).
in the "automatic"
effects of spell or amulet-not
showing any suror
even
when
do
not
prise
disappointment
they
yield the intended
have
in
And
it
has
been
result-than
they
prayer.
argued that there
is no essential or generally
valid distinction
between
spell and
prayer, since they sometimes exploit similar formulae in similar circumstances.' I
Once we have arrived at this point it is only a step to the final
defended by some scholars, and first by R.R. Marett,
conclusion,
that any distinction between religion and magic is an illusion based
on a variety of fallacies, especially ethnocentric
and
projection
historical distortion.$ The modern concept of magic, so these critics
reason, is the product of an evolution which started in late antiquity
in the context of Jewish-Christian
conflict with remnants of pagan
under
cult and which acquired its definitive Western connotations
the double influence of a comparable
theological conflict between
in the 16th century
Protestants
and Roman
Catholics
and the
of
scientific
evolution
Western
ideas.
subsequent
Magic and its
above-listed
are
an
modern-Western,
symptoms
essentially
rationalistic
and biased concept and as such quite unsuitable
for
to the study of non-Western
cultures where similar
application
dichotomies
cannot
be
either
terdemonstrated,
always
derived
or
Such
criticism,
moreover,
minologically
conceptually.
considerable
support from the discovery that in addition to all other
similarities
between magic and religion, they also could excercise
identical functions on the level of social psychology e.g. as conductors of socially unmanageable
emotions or frustrations.
since
Marett
have
Consequently,
many
anthropologists
expressed serious doubts as to the utility or relevance of the distinction and some have even advised that the terms 'magic'
and
should
be
It
is
avoided
however,
'religion'
altogether.9
noteworthy,
that some of them, among whom the most influential,
accepted this
surrender
only because, and as long as, the existing definitions did
not work out or were being disputed. Cl. Kluckhohn
(o.c. n. 9), for

181
does feel that the terms 'religion'
and 'magic'
example,
may be
contrasted
at
the
common-sense
level, just as the
meaningfully
terms "boy, youth, man, old man, which neither physiologists
nor
will wholly discard but which they will also not
psychologists
units and basic concepts
attempt to include among the elementary
which
rear
their
sciences."
upon
they
The result is a confusing spectrum of divergent theories between
the extremes of a Tylorian dichotomy between theistic religion and
and
magic seen as mutually exlusive and incompatible
concepts'
the view that magic-if
at all from religion-is
distinguishable
merely an aspect of it. In that case magic is not to be contrasted
with or opposed to other
with religion itself but to be compared
of religion,
for instance
or sacrifice."
components
prayer
between those who, enthusiastically
or reluctantly,
Significantly,
continue to adhere to some kind of a definitional
distinction
either
between magic and religion, or between magic and other aspects of
can be noted:
religion, some striking correspondences
to single out manipulative1) there is a marked
tendency
as the essential
coercive versus emotional-supplicatory
attitudes
distinction
between magic and (other components
of) religion,12
viewed in
is, since Ruth Benedict,
generally
2) the contrast
terms of theoretical
extremes
as conceptual
tools to
functioning
analyse a reality that, of course, will seldom or never answer to
these ideal-typical "abstractions".
instead,
Reality,
generally
3
a
"continuum"
between
the
two
extremes!3
displays
start from their own concept of magic,
3) all the theoreticians
whether in order to defend or refute its correctness,
relevance,
with
applicability,
general validity, ethnocentricity,
(dis)similarity
etc.
etc.
their
without,
however,
religion
always making explicit
own definition. Practically no one escapes moments of reduced concentration when they suddenly fall into using unsophisticated
common sense concepts, though they sometimes betray their awareness
of the lapse by putting the term magic between inverted commas
or adding "so-called".
One problem is that you cannot talk about
without
the
term magic. Using a term entails having
magic
using
a concept, even if to reject its applicability.
Another problem is that
magic is not always magic.

182
A field-test:

the Graeco-Roman world

Let us illustrate the latter statements

with a few examples from
the discussion
on magic in antiquity.
In antiquity
the term
Persian
magoslrnagus had a double
meaning:
priest (originally
to the tribe of the Magi) and enchanter,
wizard, with
belonging
overtones:
The
charlatan,
predominantly
negative
imposter}4
former and original one is responsible for the fact that the negative
connotations
associated
with the latter could not exercise
an
exotic and
unrestrained
tyranny.
Partly because of the strange,
shady rituals of the Persian fire-cult the term came to usurp connotations
of existing Greek terms for sorcery etc. such as gois,
which nourished
connotations
as
sorcerer,
magician,
negative
well.15 In the early imperial period, however, the negative aspects
of magus, magicus, magia and other terms belonging to the complex
of magic and sorcery underwent
significant changes. In the official
the
whole
was associated or identified
language
"magic complex"
un-Roman
and
with-and
as
just
eagerly persecuted as-indecent,
for which the usual terms were
currents,
suspected
religious
superstitio, prava religio, vel sim. Quite often, for instance, we cannot
whether
were persecuted
on the charge of
Chaldei
distinguish
or
because
advocated
convictions
which
they
religious
sorcery,
for
whatever
not
the
Roman
officials.
reason,
were,
appreciated
by
In fact, the two could obviously be regarded as identical whenever
this was useful either socially of politically,
as has been demonstrated by P. Brown in his innovative,
albeit somewhat overdrawn
of the accusations
treatment
against sorcerers in late antiquity. 16
the
theories of Durkheim
and Mauss
Combining
sociological
with Merton's
theories
on social deviance,
scholars like J.Z.
Smith17 have, for this period of antiquity,
adopted a strictly social
definition of magic, not marked by a specific type of attitude or
as an instrument
of social disaction or object, but exclusively
A
of
reclassification
made
had
qualification.
process
magic an
assemblage of deviant and illegitimate beliefs and related practices.
some sectors of society-defined
as "practices
Magic was now-in
which
were impious,
demonic
or fraudulent"
and
heretic,
"whether or not a particular idea or attitude was said to be magical
of the label
depended mainly on who said it, and the persuasiveness
depended mainly on the weight of authority behind it". These are

183
both quite apposite quotations
from the observations
of Hildred
Geertz on Tudor and Stuart England where a similar redefinition
took place. In the words of Phillips:
"The category 'magic',
then,
has utility for the study of Roman religion, but only in the context
of the person employing
it"}8
has been and proves still to be
This sociological
approach
for
our
of certain patterns of confruitful
interpretation
extremely
duct in Roman times. It allows us to understand
why "magical"
practices such as for instance miraculous
healings could be either
lauded as testimonies
of the power of the god in whose name they
were performed,
or meet with suspicion, persecution
and condemnation if regarded
as manifestations
of prava religio, inspired by
one of the problems of miracle
demonic powers,19 Consequently,
workers was to convince the world that they were not magi but took
their inspiration from "legitimate"
sources.2 Apollonius of Tyana,
the famous miracle worker of the first century A.D., found it
the language of birds
necessary to explain that he could understand
or predict earth-quakes,
not because he was a magus but simply
because he lived on a diet that made his body hyper-sensitive
to
of the air. Jesus had to defend himself against the
vibrations
that his power to do miracles derived from demonic
imputation
forces. Pagans in the first three centuries A.D. accused Christians
of magical practices (next to incest, canibalism
and atheism) and
all
in
the
and
fifth
when
it
fourth
repaid
century,
they were congot
In sum, accusations
of
fronted with the very same accusations.21
magic could function as a social and political weapon to diabolize
and eliminate the opponent.
the
Yet this social function of magic by no means precludes
of
the
term.
existence of more concrete 'substantive'
implications
As I shall argue infra, apart from an aspect of social illegitimacy
associated with magical
there were also more formal characteristics
practice. If one really wishes literary support for this truism there
is the accusation against Apuleius to prove it.22 This is granted for
instance by a faithful follower of J. Z. Smith in one of the best short
accounts of this issue: D.E. Aune, "Magic in Early Christianity"
approach, he writes:
(o.c. n. 16). In keeping with the functionalistic
is defined as that form of religious deviance
whereby
"Magic
or social goals are sought by means alternate
to those
individual

184
And
normally sanctioned
by the dominant
religious institution".
he regards magic as a species of the genus religion: "Magic
appears
is of
to be as universal a feature of religion as deviant behaviour
human societies".
In the meantime we observe that the expression
or social goals" and the term "means"
une"seeking individual
refer
of
to
'old-fashioned'
definitions
that
most
quivocally
magic
more dogmatic disciples of the social interpretation
would eat their
tongues to avoid. He even rubs it in: "goals
sought within the context of religious deviance are magical when attained through the
of supernatural
management
powers in such a way that results are
3
virtually guaranteed" .23
Cutting knots
So we are again confronted
with an embarrassing
mixture of
essentialist definitions and functional alternatives.
And this in fact
seems to be unavoidable.
Both-and
as
perhaps other approaches
well-will
have to co-operate if we do not wish simply to abolish all
communication
on concepts of magic and religion.
In order to
create at least some sort of common platform on which scholars can
if not
stage their ideas in such a way that others may understand,
their ideas on central issues, I very succintly and proviappreciate,
a few reflections.
sionally formulate
who
Scholars
deny or minimize the relevance of distinctions
1)
between magic and religion (or the relevance of the mere concept
of 'magic')
cultures where similar
usually refer to non-Western
distinctions
do not seem to operate as clearly as they are supposed
to do according
to. modern
Western
theories.
In so doing they
follow the "emic"
"first
(` `inside",
person",
"empathic")
more or less dogmatically
approach many modern anthropologists
A Diary in the
adhere to.24 However,
Malinowski's
"testament"
Strict Sense of the Term (1967) has done much to shake the foundations
of this belief. "The myth of the field-worker,
perfectly self-tuned to
his exotic surroundings,
a walking miracle
of empathy,
tact,
and
was
demolished
patience,
cosmopolitanism,
by the man who
had perhaps done most to create it. "25 Independently,
however,
or sociologist had already ventilated doubts
many an anthropologist
as to the possibility (if not the desirability)
of such an "empathic"

185
Goode reproaches
his opponents
for failing to underapproach.
stand what a concept is for: "If the natives do not 'see' a distinction
we believe scientifically
we do not discard it on that
important,
since
"we
are
in
account",
engaged
precisely that task, to create
scientific theory and concepts with which to understand
all social
Less doctrinaire
but no less clear, Evans-Pritchard
processes".
says
that "terms are only labels which help us sort out facts of the same
kind from facts which are different or in some respects different. If
the labels do not prove helpful we can discard them. The facts will
be the same without their labels".
Finally, a sociologist who has
become a professor of Ancient History, K. Hopkins writes: "Yet
as moderns and as historians we have no alternative
but to use our
own concepts
and categories
to describe
and explain
other
societies. "26 After this random list of quotations
it will come as no
statement:
surprise that I would subscribe to Snoek's categorical
"It should be clear then, that an emic scholarly approach is a contradictio in term in is. Scholarly discourse is always etic and should
therefore be conducted
in etic terms" . 2'
it
to inquire whether
Consequently,
may be more rewarding
non-Western
cultures do or do not recognize a distinction between
we
introduce and, if they do not, ask why not (and why
categories
we do), than discard apriori our own conceptual tools, a psychological tour de force which many scholars believe to be an illusion in the
first place.
to non2) But what if the categories are not only inapplicable
Western cultures but also very dubious as categories in themselves
since the various items appear to blur and are certainly not all or
not always equally distinctive? Then there are two options. One is
to avoid both the use of the term magic and any query
categorically
which entails the application
of this term even in books which have
as
their
use of more
magic
very subject. This means consistent
restricted terms as "amulets,
defixio, prayer, sacrifice, malevolent
etc.
as
as possible), without
practices",
(after
precise definitions
whereas the use of
classifying them into a manageable
taxonomy,
terms like "spell, sorcery" already becomes dangerously
risky. All
this will, of course, soon appear to be utterly unpractical.
The only
realistic alternative
is to devise at least a working definition of the
concept you are going to employ. This definition might belong to

186
the type of classifications without sharp borders, more especially the
so-called polythetic
classes, as they were for the first time forHe uses the example of the concept
mulated by Wittgenstein. Z8
he
describes
as
a family to which all games belong,
which
'game',
however,
without,
They are
sharing all the family resemblances.
and
network of similarities
linked by "a complicated
overlapping
has
been
to
the
use
of
the
conThis
criss-crossing".
insight
applied
of course, similar definitional
problems
cept of religion-where,
loom up. W.P. Alston,29 for instance
lists what he calls nine
characteristics"
and he states that "When
"religion-making
characteristics
are
to a sufficient degree, we
of
these
present
enough
have religion".
He also says that "the best way to explain the concept of religion is to elaborate in detail the relevant features of an
ideally clear case of religion and then indicate the respects in which
less clear cases can differ from this, without hoping to find any
sharp line dividing religion from nonreligion"
(ibid. ).30
practices or expressions
may share
Just like religions, 'magical'
some though not all family resemblances.
This means that we may
or prototypical'
definition
of magic,
accept a 'broad,
polythetic
bases on a "common
sense" collection of features, which may or
to convention
and experience,
may not, according
largely correto
the
items
listed
in
the
first
of
this
introduction:
part
spond
instrumental,
mechanical,
coercive,
manipulative,
non-personal,
with short-term,
concrete
and often individual
goals etc., and
coined by our
standard,
employ this as a provisional
ideal-typical
and just see what happens.
As Kluckhohn
has
cultural universe,
well observed,31 "No definition
can hope to incorporate
(...) all
established
in
the
various
fields
of
learnof
each
aspects
conception
and
remain
serviceable.
of
a
definition
ing
yet
(...) Construction
must depend upon convenience
(....) [which] demands doing as little violence as possible to whatever established core of meaning may
and scholar terexist in familiar
language
usage in ordinary
far
is consistent with
also
so
as
this
It
minology.
requires simplicity
precision .
status,
3) Any definition, given its provisional and experimental
needs
continuous
It may for instance
readjustment.
prove
"coercive"
element
to
substitute
the
preferable
(or clarify)
by the
attractive
of
action
introduced
by
"performative"
very
category

187
Tambiah.32
Certain elements included in earlier definitions
have
been ousted, definitively it seems, such as the axiomatic supposition
of the setting of magic in lower layers of society, or the presumed
lesser degree of emotion involved. Others may follow. The definition should remain open, never be taken absolutely
nor exclude
insertion
of
other
the
characteristics.
apriori
Though
purposeoriented classifications
may turn out to be of little value in one sort
of rites, this should not necessarily lead to a general denunciation.
Nor is the relevance of a classification
according to the nature of a
ritual action and the concomitant
attitudes per se affected by the
of social or socio-psychological
functions
such as the
discovery
canalization
of aggression or anxiety or the control of social tension.
Rather are there serious reasons to doubt the suitability
of funcas tools for understanding
tional elements-however
important
social processes-as
instruments
for a general classificatory
system.
This may appear from the fact that practically no scholar using the
functionalistic
approach manages to elaborate his social interpretation of magic without using the term 'magic' in one or the other of
the more conventional
substantive meanings, as we have seen in the
above.
examples
should be drawn between
4) The question whether distinctions
and
or
magic
religion
magic and other features within religion is
then, besides being a matter of personal viewpoint and, indeed, of
What is important is to make a distincbelief, of minor importance.
tion between magic and non-magic,
and it will be impossible-and,
if possible, utterly impractical-to
completely eliminate religion as
one obvious model of contrast, if only for the fact that both religion
and magic basically refer to 'supernatural'
powers.33 The question
whether one of these contrasts is religion or the propitiatory
and
within
is
of
In
supplicative
aspects
religion
secondary importance.
order to prevent misunderstanding,
one should make one's standpoint explicit and, before doing so realize that one's choice does
have considerable
When A.D. Nock, o.c. (n. 14)
consequences.
notes
on
ancient
Greece
"there
is not, then, as with us (my
312,
italics H.S.V.),
a sphere of magic in contrast
to the sphere of
one
but
religion"
may accept or dispute this pronouncement,
should at least realize that it could only be made because of the supposition implied in the italicized part.

188
Once more the Graeco-Roman

World

Ancient Greece was a non-Western

culture. Ancient Greece was
the first Western culture. Both statements
can be defended. There
is no reason whatsoever
to deny the applicability
of
dogmatically
modern
notions of magic to particular
sections of Greek ritual
that the modern
action. Moreover,
those who contend
idea of
was
coined
a
magic
originally
by
Jewish-Christian
setting in lateare
themselves
since they
antiquity
guilty of a serious reduction,
was anticipated
forget that this type of "associative
engineering"
reclassification
of the notion of magic by Roman
by a comparable
officials against Christians
related
(as
above) and, what is more,
that similar manipulations,
less
though
systematically
applied, were
as old as Greek democracy:
it was so usual to identify (undesirable)
of a new religion with magicians
and charlatans
that
prophets
in
could
with
it
the
Bacchae
the
the
of
Euripides
play
prophet
calling
a
and
a
Ninos
was
Dionysiac
disreputable
priestess
religion
goes,
new rites
convicted in the fourth century BC for having introduced
and having practised magic.34 Indeed, Greece had it all: the use of
and all sorts of material means,
spells, magical curses, incantations
to the recipes of professional
fabricated
in
according
specialists,
order to attain direct and immediate
with
the
or
without
goals,
enforced assistance of supernatural
beings. Not only did Greek have
it, it was also aware of it, reflected on it, frequently expressed disapproval of it, linked it with exotic creeds, and* used it as an effective
with rivals in love, commerce,
lawsuits
weapon in the competition
or sports.
Greece also had something
which no modern
scholar would
hesitate to call religion. We thus perceive distinctions that resemble
those drawn in the first part of this paper. And we can also see the
at other
of these distinctions,
sometimes
incidentally
blurrings
times systematically,
even consciously.
of
Conscious
blurring
the
for
is
one
of
most
characteristic
and
distinctions,
instance,
in the Greek magical papyri of late antiinteresting
phenomena
quity. Again one has a choice between several options. One may
because one denies on principle
totally ignore the phenomenon,
between
and
distinction
magic
religion. But then one should
any
quit speaking of "magical
papyri" (and religious elements in them)
in the first place. Or one can use the phenomena
of late antiquity

189
in order to support a more general theory that these categories were
blurred throughout
However, one can also maintain the
antiquity.
"ideal"
extremes
and observe that the "path
from magic to
could
be
seen as continuous
in the papyri"
as A.E. Segal
religion
o.c. (n. 16) phrases it. Personally,
I would prefer the later option.
If one starts from the idea that magic is
religion, one loses the
chance of saying that "magic was religion",
as Segal concludes for
this specific genre and period, and of explaining why such blurring
was more radical and pervasive in this cultural context than at other
times and places in the Ancient World. In my view this would be
a serious draw-back.
And if we now, just for a moment, equipped with some random
of magic (versus religion) investigate the Greek
family resemblances
and Roman evidence, do we indeed find what practically all recent
claim to find: namely that our division did not have
investigators
the same relevance in antiquity,
indeed, that it did not exist at all?
Let us follow one of them, H. Remus in a solid and useful recent
article: "Magic
or Miracle? Some Second Century
Instances".35
The author inquires on what grounds the negative labels magos,
magikos etc. were applied. He analyses all those aspects that we commonly associate with the label magic as we have listed them above:
of secret texts, comcharms, gestures, application
manipulations,
pulsory action through the use of secret names of gods or demons
etc. etc. And of course he finds that using a powerful name may be
labelled magic in one case and religion in the other depending
on
the position of the god concerned in the value system of the dominant group. The same is true for objects: an amulet of a pagan god
will be listed among magic, whereas the use of a crucifix for miracle
cures may be accepted as a deed of religious fervour. Indeed: "your
.36 All the same the author has
magic is my miracle, and vice-versa"
to concede "that there is some agreement
that magic involves comand he produces
a number
of unequivocal
pulsion"
quite
statements
in which the supplicatory
attitude of prayer is radically
one of magical spells. Nonetheless
it is
opposed to the compulsory
also true that "the canons were not tight and the texts we have
examined
were not made by impartial
suggest that judgments
on
other
but
to
the
detriment
of 'their' rites and
testing
grounds,

190
and the advantage
of 'ours' ". So he concludes,
as we
expect by now: "In these examples it is evident that manipulations
as such do not really demarcate miracle from magic, except for perbeforehand
to make that demarcation.
Such detersons determined
which
mination
roots in social and cultural
are more
settings,
in
than
are
the
such conflicts
important
understanding
manipulations which are found both in magic and in miracle".
of his
Now, I do not for one moment dispute the correctness
it
realized
that
should
be
However,
"your magic is my
analysis.
where
miracle"
is no less true for modern-Western
perception,
to recent views the substantivistic
distinction
between
according
bottle of a magical
magic and religion came to existence .3 A
medicine sold by a mysterious
foreigner at the fair is often opposed
to the bottle from Lourdes, which is religiously sanctioned,
though
both are chemically the same and are used for the very same purin assesposes. However, my point is that this social differentiation
ment, true and important as it may be, as such by no means entails
the necessity to reject the concept magic for use in comparative
Such inferences could be considered
research.
only if charges of
ancient
and
were
devoid
modern,
of any concrete
consistently
magic,
details or argumentation. However, these charges were generally buttressed with references
to all sorts of suspicious
practices,
questionable gods and demons, secrecy and so on.
the ancients no less than modern people knew what
Apparently,
would come in handy to underpin their charges.
specific imputations
The questions of whether these charges had any basis in real practice and whether the materials might also function on the right side
of the borderline
between religion and magic, though interesting
from a different perspective,
are of no direct importance
for the
issue under discussion. What concerns me now is the perception of
the deviant forms of religion or outside religion and the practices
they were consistently associated with. And it is this ancient representation of the magical attitude which often closely corresponds
with
of Tyana
what we call magic. As we noticed earlier Apollonius
the language
found it necessary to explain that he could understand
of birds or predict earth-quakes,
not because he was a magus but
simply because he lived on a diet that made his body hypersensitive to vibrations of the air. Also, the authors or revisors of the

wonders

191
that
to eliminate
have done wonders
Testament
anything
the use of saliva for the
might recall sorcery in Jesus' behaviour,
dumb
and
of
a
deaf
apparently
being an isolated slip
person
healing
of the pen.3$ Now, if Jesus or his biographers
systematically
which might arouse suspicion
avoided references to manipulations
of magic, they must have known what to avoid. In the life of
Theodorus
it is told that the blessing of this holy man, and no
an
amulet
by a wise woman, was what was now
longer
prepared
from
the effects of a green lizard that had
to
protect you
supposed
New

fallen into your soup.39 Of course, this is also an instance of "your

But to exclude other approaches
of intermagic is my miracle".
the
same
would
be
reductionist.
For
pretation
seriously
exactly
author of Sacred Disease ( 15th
distinction is made by the Hippocratic
c. BC) who opposes the use of katharmoi and epoidai (purificatory
medicines and spells), which he denounces
because they are used
to compel the gods, to thuein, euchesthai and hiketeuein tous theous
(sacrifice, pray and supplicate the gods), which he approves of as
examples of a pious attitude.4 All this means that apart from the
there were also more formal characterisaspect of social judgment
tics
generally associated with magic action.41
To conclude. I was enticed into treading this slippery path by a
research project in the field of ancient religion, in casu an analysis
of the complete
corpus of the so-called defixiones from GraecoRoman
These
short
curse-texts,
antiquity.42
mostly
usually
to harm or bind a person
inscribed in lead tablets, are intended
does
whom the author, mostly for reasons of envy or competition,
not like. Starting from basic formulas such as: "I bind or fix or nail
down John, that he may not move, breathe, speak" etc., the followcan be found: a god or demon from the
ing common characteristics
Netherworld
is invoked, who is ordered to execute the wish of the
the author is
without exception
author of the tablet. Practically
and there is no legitimation
of the harmful act. Often
anonymous,
the tablet is transfixed with a needle. Also a kind of voodoo-doll
may be added.
During my research I found a number of texts that had always
been included in the corpora of defixiones, but which nevertheless
a different atmosphere
and phrasing.
Here we read for
betrayed

192
has stolen my coat (or: I suspect John of the theft).
oh mighty Lord, I implore you: come to my aid,
punish him, that he may suffer (or in other texts: that he may
return my property).
Please hear my prayer, 0 Lord, for I, Peter,
have been wronged".
There are clear and rather consistent
differences with the common type of defixio: the author makes himself
known, he has every right to his action for he has been injured and
there is generally
an argument
or legitimation
for the
therefore
instead of demons
action; instead of pressure there is supplication,
we see great gods of the heavens, Zeus,
or gods of the Netherworld
the Sun, Theos Hupsistos.
I have baptized
the latter, clearly deviant
types of defixiones
or
for
Of
course, there are also
'judicial prayers'
'prayers
justice'.
mixtures of both types, but my point is that the differences between
the ideal type of the defixio proper and that of the judicial prayer
that former
appear to correspond
strikingly with the distinctions
used
to
the
associate
with
between
generations
opposition
magic
and religion. So, for the time being I shall continue to label my
defixiones magical acts (in which I follow the common practice even
definitions of magic)
adopted by those who reject the conventional
and my judicial prayers religious acts, and I feel supported by the
fact that here at least the ancient authors display an unequivocal
and explicit awareness of the differences.
And I believe it may be
useful to investigate whether we detect these distinctions
based on
in
characteristics
other
sectors
too.
We
can
never
do so
practical
without the heuristic
tools provided
our
'broad'
definitions.
by
Whether this will turn out to be as useful for other cultures as it is
for the Greek and Roman world I do not know. Perhaps the Greeks
were prototypically
biased rationalists
avant la letmodern-Western
tre ? But if we begin by rejecting terms, concepts and definitions
right from the outset, nobody will ever know.
instance: "John
If I am right,

88
Herenweg
NL-2361 EV Warmond

H.S.

VERSNEL

1 I am
greatly indebted to Richard Gordon, who read an earlier draft of this
paper, emended the English, offered a shower of generous critical observations
and who, even after the many alterations prompted by his criticism, will certainly
not approve of the result. As I have added substantial parts in a later stage nobody
but the author can be held responsible for form and content of the final product.

193
2 Inter alia for the reason that the existence of notions of religion and magic in
antiquity is hardly debatable, while a concept of science is patently absent. See
Phillips o.c. (below n. 16) 2700 ff. Modern discussion on the relationship magicscience :K.E. Rosengren, "Malinowski's Magic: The Riddle of the Empty Cell",
CurrentAnthropology17 (1976) 667-685 and some of the comments. Note that since
J. Goody, "Religion and Ritual: The Definitional Problems", BritishJournal of
Sociology12 (1961) 142-157, many scholars tend to view the dichotomy profanesacred (or: supernatural) as a Western ethnocentric construct.
3
W.J. Goode, "Magic and Religion. A Continuum", Ethnos 14 (1949) 172182 ; idem, Religionamongthe Primitives(Glencoe 1951).
4 This, in fact, is a summary of Malinowski's famous theory as presented in:
"Fishing in the Trobriand Islands", Man 53 (1918) 87-92, and elaborated in:
Magic, Scienceand Religion(1925) and the article "Culture" in: Encyclopediaof the
SocialSciences4 (1931) 634-42.
5 In: J. Gould and W.L. Kolb, Dictionaryof theSocialSciences(Glencoe 1964) s.v.
magic. Cf. ibidem"Magic is the use of 'dogmatic knowledge' for control, religion
the worship or propitiation of beings known dogmatically." More extensively R.
Benedict had said much the same in her sensible article "Magic" in Encyclopedia
of theSocialSciencesX (1933) 39 ff. It should be noticed that Leach recanted in later
works.
6 See: K. Thomas, Religion and the Decline of Magic (Harmondsworth
1973 = 1971). The exchange with Hildred Geertz, in: "An Anthropology of
Religion and Magic", Journal of InterdisciplinaryHistory 6 (1975-6) 71-108, well
illustrates the confusions which result from the adoption of diametrically
opposite-substantive/essentialist versusfunctional-definitions.
7 See especially
J. van Baal, "Magic as a Religious Phenomenon", Higher
Educationand Researchin the Netherlands7 (1963) 10-21; W.E.A. van Beek, "The
Religion of Everyday Life: An Ethnoscience Investigation into the Concepts of
Religion and Magic", in: W.E.A. van Beek and J.H. Scherer, Explorationsin the
Anthropologyof Religion.Essays in Honourof J. van Baal (The Hague 1975) 55-69. H.
Philosooph, "Primitive Magic and Mana", Man 6 (1971) 182-203, argues that
magical action is often employed in order to influence autonomous (divine)
powers.
8 I refer for a
survey of this discussion and the bibliography to M. and R. Wax,
"The Notion of Magic", CurrentAnthropology4 (1963) 495-518; more pointedly:
Dorothy Hammond, "Magic: A Problem in Semantics", AmericanAnthropologist
72 (1970) 1349-1356. Cf. Rosengren o.c. above n. 2, and the comments by
referees. H.G. Kippenberg and B. Luchesi (edd.), Magie, Die sozialwissenschaftliche
Denkens(Frankfurt 1978) provides a useful colKontroverseberdas Verstehen fremden
lection and a good introduction by Kippenberg.
9 A.R. Radcliffe-Brown, Structureand Functionin PrimitiveSociety
(Glencoe 1952)
138: ".... to avoid as far as possible the use of the terms in question until there
is some general agreement about them." C. Kluckhohn, in: A.L. Kroeber et alii
(edd.), AnthropologyToday (Chicago 1953) 518: "Anyone can make a definition
that will separate magic from religion; but no one has yet found a definition that
all other students accept"; D.F. Pocock, "Foreword" to Marcel Mauss [transl.
R. Brain] A GeneralTheoryof Magic (London 1972) 2: "If categorical distinctions
of the Western mind are found upon examination to impose distinctions upon (and
so falsify) the intellectual universes of other cultures then they must be discarded
or, as I have put it, dissolved. I believe 'magic' to be one such theory ..."

194
10 R. Horton, "A Definition of Religion and its Uses", Journal of the
Royal
AnthropologicalInstitute 90 (1960) 201-220: religion entails a belief in spiritual
beings, magic is a "secular" phenomenon; M. Spiro, "Religion: Problems of
Definition and Explanation", in: Anthropological
Approachesto the Study of Religion,
A.S.A. Monographs 3 (1966) 85-126: magic like politics or art is a functional alternative to religion. Cf. also E.E. Evans-Pritchard, Theoriesof Primitive Religion
(Oxford 1965) 33. Most of this in the context of a revival of Tylor's 'substantivistic' approach of religion. See: R. Horton, "Neo-Tylorianism: Sound Sense
or Sinister Prejudice?", Man 3 (1968) 625-634; E. Ross, "Neo-Tylorianism: A
Reassessment", Man 6 (1971) 105-116.
11 This is the conclusion of D. Hammond o.c.
(above n.8) 1355, "Magic would
serve as a denotation for one type of ritual behavior (viz. "mechanical religious
practices") and thus belong to the set of terms in that category, such as prayer
or sacrifice. With these propitiatory rites it is contrastive, not with religion",
somewhat along the lines drawn by E. Norbeck, Religionin PrimitiveSociety(New
York 1961). Quite close to this: J. van Baal (o.c. above n. 7), who defines magic
as:"all those acts and spells directed towards furthering a certain aim by employing another reality than the empirically determinable one, in which process this
other reality is not independently active but is an instrument manipulated by the
active person", and who sets magic within the category religion.
12 For instance: R. Benedict, o.c.
(above n. 5); R.H. Lowie, History of
EthnologicalTheory(New York 1937) 103; idem, PrimitiveReligion(New York 1948)
141; E. Norbeck, Religionin PrimitiveSociety(New York 1961) and many others.
See the survey by D. Hammond o.c. (above n. 8) and her conclusions (p. 1350):
"These two characteristics-impersonal
forces and manipulative techniquesremain as current critical diagnostic features." Note that the same characterization of magic is accepted by D. Hammond herself and J. van Baal (preceding
note), who both denied the dichotomy magic-religion.
13 So in nucealready R.R. Marett, The Threshold
of Religion(London 1914) 28,
but much more cogently formulated by R. Benedict o.c. (above n. 5) 40; eadem,
"Religion", in: F. Boas (ed.) GeneralAnthropology(Boston 1938) 647; the term
"continuum" was made popular by Goode o.c. above n. 3.
14 See on the development of the terms: A.D. Nock, "Paul and the
Magus",
in: idem, Essays on Religionand the Ancient WorldI (Oxford 1972) 308-330; G. Delling, in: TheologischesWrterbuchzum Neuen TestamentIV, 360-363.
15 See: W. Burkert, "Gos", RM 102 (1962) 36-55.
16 P. Brown, "Sorcery. Demons and the Rise of Christianity: from Late Antiquity into the Middle Ages", in: idem,Religionand Societyin the Ageof Saint Augustine
(London 1972) 119-146. On these processes also: R. MacMullen, Enemiesof the
Roman Order (Cambridge, Mass. 19752) 95-127. Important reflections in: D.E.
Aune, "Magic in Early Christianity", ANRW II. 23.2 (1980) 1507-1557: A.F.
Segal, "Hellenistic Magic: Some Questions of Definition", in: R. van den Broek
and M.J. Vermaseren (edd.), Studies in Gnosticismand HellenisticReligionsfor G.
Quispel (Leiden 1981) 349-375, with honest criticism of some of P. Brown's
theories. Most recently and most important: C.R. Phillips III, "The Sociology of
Religious Knowledge in the Roman Empire to A.D. 284", ANRW II, 16.3
(1986), 2677-2773, espec. 2711-2732. Cf. also below nn. 19 and 20.
17
J.Z. Smith, "Good News is no News", in: idem, Map is not Territory.Studies
in the History of Religions(Leiden 1978) 190-207.
18 H. Geertz, o.c.
(above n. 6) 75; Phillips o.c. (above n. 16) 2718. Cf. ibid.

195
2579: "Terms such as 'magic' or 'orthodoxy' (...) represent conceptual models of
the group which employs them: they do not possess absolute, eternal authority",
and: "the appearance of a term like 'magic' represents a value-judgment of the
author employing it, whether the author be Apuleius, Origen, Sir James Frazer
or Arthur Darby Nock".
19 Most recently on the strategies applied in sorting out approved from disapproved practices and in attributing them to 'true religion' and 'magic' respectively : J. Neusner, E.S. Frerichs, and P.V. Flesher (edd.), Religion, Science,and
Magic in Concertand Conflict(Oxford 1989), who rightly contend that by studying
these processes we gain insight into the groups' self-definition. The principle had
been recognized long ago; L. Bieler, TheiosAner (Darmstadt 1967 = Wien I-II,
1935-6) 84-87, who already refers to G.P. von Wetter, 'Der SohnGottes'.Eine Untersuchungber denCharakterund die Tendenzendes Johannes-Evangeliums
(Gttingen 1916)
vonEpidauros (PhilologusSuppl. 22/3, 1931)
73-82; R. Herzog, Die Wunderheilungen
cf. also: G. Theissen,
140; "Aberglaube ist immer der Glaube der Anderen";
"
UrchristlicheWundergeschichten
(Gtersloh 1974), 230: 'Aberglaube' ist dann der
in einer Gesellschaft abgelehnte Glaube, 'Glaube', so knnte man ironisch formulieren, der offiziell anerkannte Aberglaube. Wo die Grenze zu ziehen ist,
bestimmen die massgeblichen Kreise" (as quoted by Remus 1983, o.c. [next note]
207 n. 2).
20 Anitra B. Kolenkow, "A Problem of Power: How Miracle Doers counter
Charges of Magic in the Hellenistic World", Societyof Biblical Literature. Seminar
o.c. (above n. 16) 356 f.; Phillips o.c.
Papers (Missoula 1976) 105-110. Cf. Segal
"
'Magic or Miracle'? Some Second Cen(above n. 16) 2716 n. 112; H. Remus,
tury Instances", The SecondCentury.A Journal of EarlyChristianStudies2 (1982) 12756, reissued as ch. IV of his book: Pagan-ChristianConflictoverMiracle in the Second
Century(Cambridge Mass. 1983).
21 See the
article "Heidenverfolgung", in RAC; Phillips o.c. (above n. 16) 2713
n. 104; R. Turcan, "Les motivations de l'intolrance chrtienne et la fin du
mithriacisme au IVe sicle ap. J.-C.", Actes VI Congr. FIEC II, 209-226.
22 See the discussion by A. Abt, Die Apologiedes Apuleius von Madaura und die
antikeZauberei(Giessen 1908). Cf. also: N. Fick, "La magie dans les Metamorphoses
d'Apule", REL 63 (1985) 132-147.
23 With the observation: "those who failed to achieve the
particular benefits
promised by religion not infrequently turned to magic for an alternate means and
one generally regarded as socially deviant for achieving their goals", he differentiates between the substantive and the functional characteristics, but adopts both
as relevant.
24 For fundamental discussion see: P. Winch, Die Idee der Sozialwissenschaft
und
ihr Verhltniszur Philosophie(Frankfurt 1974); A. MacIntyre. "A Mistake about
Causality in Social Sciences", in: Philosophy,Politicsand Society2nd ser. (edited by
P. Laslett and W. Runciman, Oxford 1967) 48-70; I.C. Jarvie, The Revolutionin
Anthropology
(London 19672);J. Agassi and I.C. Jarvie, "Magic and Rationality",
24 (1933) 236-45, some of which are collected in German
British Journalof Sociology
translation in: Kippenberg and Luchesi o.c. (above n. 8).
25 Cl. Geertz, "From
the Native's Point of View'. On the Nature of
Anthropological Understanding", in: K.H. Basso and H.A. Selby, Meaning in
Anthropology
(Albuquerque 1976) 221-237, esp. 222. See on the concepts of "emic"
and "etic" the good discussion by M. Harris, The Rise of AnthropologicalTheory
(New York 1968) 568-604.

196
26 Goode in his reaction to the article of the Waxes, Current
4 (1963)
Anthropology
507; E.E. Evans-Pritchard, Witchcraft,Oraclesand Magic amongthe Azande(Oxford
1937) 11; K. Hopkins, "Contraception in the Roman Empire", Comparative
Studiesin Societyand History8 (1965) 124-157. "Wie knnen denn fremde Glaubensanschauungen beschrieben werden, ohne da der Beobachter seine eigenen
Begriffe mit heranzieht?" (Kippenberg o.c. above n. 8, p. 43, and see his
discussion).
Approachto the Application
21J.A.M. Snoek, Initiations.A Methodological
of Classification and Definition Theoryin the Study of Rituals (Diss. Leiden 1987) 7.
28 L.
Wittgenstein, PhilosophicalInvestigations(New York 1958, translated from
the German ed. 1953) I, 66-7.
29 "Religion",
Encyclopediaof Philosophy,vol 7 (1967) 142. The principle of
polythetic classification is exemplarily exploited by J.Z. Smith, "Fences and
Neighbors: Some Contours of Early Judaism", in: idem, ImaginingReligion:From
Babylonto Jones Town (Chicago 1982) 1-18. Its import for anthropological studies
has also been questioned, for instance by R. Needham, "Polythetic Classification:
Convergence and Consequences", Man 10 (1975) 349-369.
30 This is a
prefiguration of what recently has been labelled the "prototype
theory" of classification, as was shown by B. Saler in a paper read during the Congress of the IAHR at Rome, september 1990. I am indebted to this paper for some
information on problems of the definition of religion.
31 C. Kluckhohn, "Values and Value-Orientations in the
Theory of Action. An
Exploration in Definition and Classification", in: T. Parsons and E.A. Shils (eds.)
Toward a GeneralTheoryof Action (Cambridge Mass. 1959) 394/5. I am painfully
aware of the injustice I have done to the infinitely intricate problems of the definition of definitions, as one can learn from J.A.M. Snoek, o.c. (above n. 27), who,
on p. 9 f., largely advocates the strategies as formulated in my text.
32
S.J. Tambiah, "The Magical Power of Words", Man 3 (1968) 175-208; idem,
"Form and Meaning of Magical Acts. A Point of View", in: R. Horton and R.
Finnegan (edd.), Modesof Thought(London 1973) 199-229. For a variant see: J.
Skorupski, Symboland Theory.A PhilosophicalStudy of Theoriesof Religion in Social
Anthropology(Cambridge-New York 1976). Cf. also E.M. Ahern, "The Problem
of Efficacy: Strong and Weak Illocutionary Acts", Man (1979) 1-17. This
approach was made productive i.a. by Rosengren, o.c. (above n. 2), and cf. J.
Winkler and Chr. Faraone in: Chr. Faraone and D. Obbink (edd.), Magika Hiera.
Ancient GreekMagic and Religion (Oxford 1990). However, as Richard Gordon
warned me, D.S. Gardner, "Performativity in Ritual. The Mianmin Case", Man
15 (1983) 346-60, argues that the crucial features of initiation rituals, and therefore others concerned with magical or religious conceptions, cannot be understood
(exclusively) as 'illocutionary' acts: "The similarities are superficial and derive
from the putatively automatic effects of the rituals".
33 Cf.
Rosengren o.c. (above n. 2) 678: "It makes better sense to start with a
distinction between ordinary reality (or realities) and extraordinary reality (or
realities)".
34 I have collected and discussed the evidence on identifications of
magic and
foreign cults in my Ter Unus. Isis, Dionysos, Hermes: Three Studies in Henotheism
in Greekand Roman ReligionI. [Leiden 1990]) 102-123, and the case
(Inconsistencies
of the Bacchaepp. 156-205.
35 o.c.
(above n. 20).
36
According to the often quoted expression by R.M. Grant, Gnosticismand Early
Christianity(New York 1966) 93.

197
37

Consequently, a remark such as "The definitional boundaries of the term

(magic) in the ancient world for a long time remained remarkable fluid" (Phillips
o.c. [above n. 16] 2718), runs the risk of losing sight of similar vaguenesses in
modern applications of the term.
38
Jesus is often interpreted as essentially a Hellenistic type of magician, most
emphatically by M. Smith, Clementof Alexandria(Cambridge Mass. 1933) 220-237;
and Jesus the Magician (New York 1978). Once again magic is not always magic,
depending as it does upon the authority of the person who is using the term: F.
Zeit (Leipzig 1922) saw in the healing of
Preisigke, Die Gotteskraftder frhchristlichen
a woman by the touching of Jesus' clothing a pure example of magic, but W.L.
Lane, The GospelAccordingto Mark (Grand Rapids 1974), 192, says: "By an act of
sovereign will God determined to honour the woman's faith in spite of the fact(my
italics H.S.V.) that it was tinged with ideas that bordered on magic." In a similar
apologetic vein: H.C. Kee, Medicine, Miracle and Magic in New TestamentTimes
(Cambridge 1986). The two sides of the medal in: E.V. Gallagher, DivineMan or
Magician? Celsusand Originon Jesus(Chicago 1982). See on these issues especially:
Aune o.c. (above n. 16); Phillips o.c. (above n. 16) 2757 ff..
39 V. Theod. cc. 124 and 143, as quoted by P.R.L. Brown, Societyand the Holy
in Late Antiquity (Berkeley-London 1982) 151.
40 SacredDisease2, 12-13 and 4, 36 ff. Remus o.c. 145, tries to downplay them
by taking them as expressions of "adherents of one culture pitted against those of
another, with antithetical premises and values". But the opposition between
'magical' spells or charms, and pious supplicatory prayer is by no means restricted
to this text. And even so, the Hippocratic author, just like the author of the Vita
Theodori,knew what to censure.
41 A
phrase by Segal o.c. (above n. 16) 367, may help to clarify what I mean:
"Thus the charge of 'magic' helps distinguish between various groups of people
from the perspective of the speaker but does not necessarily imply any essential
difference in the actions of the participants. In a narrative about the event, the
narrator will attempt to clarify his grounds for distinction". This passage (rightly)
indicates two valid questions in the approach of magic: the functionalist one (who
speaks about whom, to whom, in what ways, and why?) and the substantivist
(exactly what is being said, in terms of adducing concrete evidence ?).
42 The results have been published in "Beyond Cursing: The Appeal to Justice
in Judicial Prayers", in: Chr. Faraone and D. Obbink (edd.), Magika Hiera.
Ancient GreekMagic and Religion(Oxford 1991), 60-106.